"The modern data stack" is the name the industry gave, in the late 2010s, to a way of building data platforms from cloud services: a managed ingestion tool copies data from sources into a cloud warehouse, SQL models transform it inside the warehouse (ELT rather than ETL), and BI, reverse ETL and machine-learning tools read the results. Each layer was a separate product, bought as a subscription and connected to the others. The approach made a working platform possible in days instead of months, and it is still the reference picture most job descriptions assume.